INDEX

Explanations

phrases indicating personal or subjective ownership or involvement

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and phrases indicating ownership or personal connection

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

é¾įåĸļå£«

-0.62

 Analytics

-0.62

irect

-0.62

icity

-0.62

 HTTPS

-0.61

 Recon

-0.60

 Disclosure

-0.60

 Returns

-0.58

 Accessed

-0.57

endo

-0.57

POSITIVE LOGITS

 fault

1.33

 thing

0.97

cup

0.96

 Fault

0.96

 liking

0.92

 problem

0.89

 usual

0.87

 style

0.84

 intention

0.80

 idea

0.80

Activations Density 0.096%

phrases indicating personal or subjective ownership or involvement

possessive pronouns and phrases indicating ownership or personal connection

No Comments

No Known Activations